Weekly Offline meeting

→ Europe/Zurich
28/S-029 (CERN)

28/S-029

CERN

35
Show room on map
Federico Carminati (CERN)
Description
Title: Weekly Offline Meeting Description: Weekly Offline Meeting Community: ALICE Meeting Access Information: - Meeting URL http://evo.caltech.edu/evoGate/koala.jnlp?meeting=M9MIMi2e2nDaDi9s99DD99 - Phone Bridge ID: 1154885 Central European Summer Time (+0200) Start 2009-07-31 14:00 End 2009-07-31 17:00 Japan Standard Time (+0900) Start 2009-07-31 21:00 End 2009-08-01 00:00 Eastern Daylight Time (-0400) Start 2009-07-31 08:00 End 2009-07-31 11:00 Pacific Daylight Time (-0700) Start 2009-07-31 05:00 End 2009-07-31 08:00 EVO Phone Bridge Telephone Numbers: --------------- - USA (Caltech, Pasadena, CA) +1 626 395 2112 - Switzerland (CERN, Geneva) +41 22 76 71400 - Slovakia (UPJS, Kosice) +421 55 234 2420 - Italy (INFN, several cities) http://server10.infn.it/video/index.php?page=telephone_numbers Enter '4000' to access the EVO bridge - Germany (DESY, Hamburg) +49 40 8998 1340 - USA (BNL, Upton, NY) +1 631 344 6100 - United Kingdom (University of Manchester) +44 161 306 6802 - Australia (ARCS) +61 0261128742 Canberra 0282124591 Sydney 0362370281 Hobart 0386858362 Melbourne 0731390705 Brisbane 0884631011 Adelaide 0864616718 Perth - Netherlands (Nikhef, Amsterdam) +31 20 7165293 Dial '2' at the prompt - Canada (TRIUMF, Vancouver) +1 604 222 7700 - Czech Republic (CESNET, Prague) +420 95 007 2386
    • 14:00 → 15:00
      Production Review 1h
      Speaker: Peter Hristov (CERN)
      Slides
      All the reported fixes will be committed to v4-17-Release. A new tag will be provided on Monday.
    • 15:00 → 16:50
      AIP-PH Coordination Meeting - Discussion on analysis readiness
      • 15:00
        Readiness of the next analysis train 15m
        M. Gheata: For what concerns the readiness for the next train, we did some test just before going to holidays: we have a new production mode where the jobs are submitted by the LPM production manager (the same software that is submitting the simulation jobs). We give to Costin a file containing the location of the jdl and of the data collections. This file is automatically generated by the alien plugin. This works fine. The things to be checked/discussed: - the output goes now to production_directory/AOD. What if we produce delta AOD's in a second step ? - I did not check the train with the trunk of AliRoot - We still run the train macro as user - we need to make a jdl that does this automatically and we just provide the train configuration file. Decision: - Early next week Mihaela will run the train with priority for the filter and vertexing. - ESD/AOD -> histos will follow after careful testing
      • 15:15
        Review of the Tags and how to update 15m
        text
        AliEventTags might not be up to date, especially for the rare processes. Furthermore. it is not clear how to take into account different quality selection cuts when counting number of tracks. Decision: - Adam Kisiel will contact the PWG conveners in order to obtain an updated list of Event tags. - Several versions of tags will be produced. Web based documentation of each version is mandatory.
      • 15:30
        Review of standard cuts used for the filter 15m
        text
        Several sets of tight and loose track quality selection cuts exist. We have to decide which cuts to use for the AOD production. Decision: Physics coordinator has to decide which cuts should be used.
      • 15:45
        Possible problems with friend chains (used for delta AODs) 15m
        A. Dainese: When creating the friend chain with the delta files (either by replacing the file name or by relying on the path), one has to check that both AOD and deltaAOD files exist in the catalogue, and only then add both to the respective chains.Otherwise you loose the correspondence. In principle this should granted by the validation, but may be not the case if the deltas are produced at a second stage. At the moment, for vertexingHF, we use the macro $ALICE_ROOT/PWG3/vertexingHF/MakeAODInputChain.C to create the two chains, checking the file existence in the catalogue. It would be very useful to have this check built in the framework(InputHandlers?). A. Morsch: If we have 2 chains A and B, and B is friend of A the two chains can get out of synch if one of the friend files is missing. The behavior is like in the following example fileA1 fileB1 events 0 - 199 fileA2 fileB2 events 200 - 399 fileA3 fileB3 events 400 - 599 Let's assume fileB2 cannot be opened. chainA->GetEntry(200) gives a message Error in <TFile::TFile>: file fileB2 does not exist and gives you back entry 400 as chainA->GetEntry(400) does The second time you call chainA->GetEntry(200) there is no error message at all. I wonder whether this is a bug or a feature. In any case this seems to be rather unsafe in the grid world. Decision: AM contacts Rene to find a solution for the TChain implementation or usage. Submitted as Savannah bug #53978
      • 16:00
        Set-up of an agile system for fast QA during shifts 15m
        Need for an agile system for QA during the run for the shifters. This would consist of fast event reconstruction on CAF and subsequent run of an analysis train or a simple (string based) selectors. Jan Fiete: Reconstruction on CAF works but output files are sent back to the user and are not kept and registered for further processing. We need a post-processing that makes a proof collection out of the output files left in the sand boxes. Extra infrastructure is needed. Develop proof storage. Decision: Jan Fiete will bring up the issue at the next CAF/PROOF meeting and assess the feasibility of such a system to be implemented in the next 2-3 months.
      • 16:15
        Calorimeter alignment information in ESD and AOD 15m
        Gustavo: I have modified the classes AliESDRun and AliESDEvent, so that the EMCAL geometry alignment class can be stored in the ESDs, following the PHOS example, I just copy pasted. See the attachments and the diff below in this message. Could you please give the OK to Cvetan and Peter so that they can commit to trunk and release? I have also prepared the modification for the EMCAL reconstructor but I will wait for your approval and also for tests that I want to do. Also I want to start the discussion about the implementation of the matrix in the AODs. Since our recalibration will run several iterations on AODs we have also to transfer it from ESDs to AODs. Cvetan proposed that we keep this matrix in the ESDRun but I don't see the equivalent for the AODs, any idea on how can we proceed? AM In principle for AOD the alignment information can be obtained from the OCDB via the Tender. Decision: - Matrices will be implemented also for AOD - A. Gheata develops a basic tender mechanism - notification of run change - connection to current OCDB entries - locking mechanism - call of detector tender tasks in a loop
      • 16:30
        Place for subsidiary files needed for analysis (CORFW output, background maps, ...) 20m
        Auxiliary files are needed for analysis and have to be available on alien. However, alien does not yet provide automatic versioning (except the trivial via file names). Decision: The auxiliary files for analysis (correction and background maps) should be stored under svn and will be available on alien. Proper versioning and documentation is the responsibility of the PWGs.